The project is divided into 3 subprojects: 1.Iterative automatic methods for refining multiple sequence alignment. Currently starting point of constructing CDD database is a multiple alignment imported form protein family models as SMART and PFAM. Since all known multiple sequence alignment algorithms are less than perfect the imported alignment have errors that are now removed by hand. The goal of the project is to provide reduce and simplify this manual work. 2.New approaches for detecting structural similarities. The goal of this subproject is to develop new approaches for fast detection of structural similarities that would complement currently used method (VAST). Thus particular attention is given to loop structure and developing of indexing methods. 3.Applications of graph theory similarity relations between domains. The similarity relation between proteins sequences or structures can be conveniently represented as a graph. The goal of this subproject is to study the properties of such graphs and develop tools for their analysis.